While the social and emotional costs of friendlessness and peer rejection are well-known, few studies have tested whether peer relationships and friendships impact youths'health by altering their eating behavior and physical activity. The proposed research is designed to assess whether peer relations and friendship can promote changes in food intake and choice of activities. Specific Aim 1 will test the hypothesis that "assortative friendship" in overweight children reinforces and maintains overeating, time allocation to sedentary behavior and social markers of victimization. Overweight and normal-weight participants will be tested on three occasions: with an unfamiliar peer, with a friend and alone. Time allocation to eating and activities and food intake will be compared across sessions. Friendship quality, peer victimization, social cognitions and internalizing symptoms will be assessed on the last session completed. We expect overweight friends to share more obesigenic behaviors such as less time allocation to active leisure activities, more time engaged in eating and greater energy consumption. We also predict similarity between overweight children and their mutual friends in the extent to which they report peer victimization and internalizing symptoms, but that a high quality friendship will diminish psychological distress. Specific Aim 2 will test whether the behavior of peers and friends may be used to modify youths'eating and time allocation to alternative activities, using a modeling paradigm. Overweight and normal-weight youths will be exposed to an informational video on healthy eating and physical activity or a control video (study skills). Then we will assess whether the child exposed to the healthy behavior video influences the eating behavior and activities of a friend or an unfamiliar peer depending on the experimental condition. It is hypothesized that exposure to the healthy video will predict food intake and choices of activities of the other child in the dyad, whereas exposure to a control video will not predict these variables. Specific Aim 3 will determine whether friends and peers can modify the reinforcing value of physical activity in overweight and lean youths, using a behavioral choice paradigm. The amount of work youth perform to gain access to food and to physical activity will be compared when physical activity is performed alone and when it is performed with a friend or with a peer depending on the experimental condition. The prediction is that overweight youth will show a biased allocation of their pattern of responses to gain access to food points when alone and when the alternative involves being physically active in the presence of an unfamiliar peer. However, the presence of a friend will increase the reinforcing value of physical activity in overweight youth. This research is designed to understand the role of peers and friends in youth adherence to health recommendations. Public Health Relevance: The proposed research is designed to assess whether peer relationships and friendship can promote changes in food intake and choice of activities. Findings from this study may improve understanding of barriers to weight loss in overweight youths and past non-adherence to health recommendations.